多光谱视网膜成像动静脉自动分割方法  

Automatic segmentation of arteries and veins in multispectral retinal imaging

作　　者：杨昆[1] 杨月婷 路宇飞 金梓 周传清 Yang Kun;Yang Yueting;Lu Yufei;Jin Zi;Zhou Chuanqing(College of Quality and Technical Supervision,Hebei University,Baoding 071002,China;College of Medical Instrument,Shanghai University of Medicine and Health Sciences,Shanghai 201318,China;Peking University Shenzhen Graduate School,Shenzhen 518055,China;Eye Hospital Affiliated to Wenzhou Medical University,Wenzhou 325027,China)

机构地区：[1]河北大学质量技术监督学院,保定071002 [2]上海健康医学院医疗器械学院,上海201318 [3]北京大学深圳研究生院,深圳518055 [4]温州医科大学附属眼视光医院,温州325027

出　　处：《电子测量技术》2023年第10期84-91,共8页Electronic Measurement Technology

基　　金：国家自然科学基金(61875123);广东省基础与应用基础研究基金(2021A1515110747);深圳市科学技术项目(1210318663);河北大学多学科交叉研究项目(DXK201914);河北大学校长科研基金(XZJJ201914);河北省大学生科技创新能力培育专项(22E50041D)资助。

摘　　要：为了解决手动标记动静脉耗时耗力的问题,开发了一种基于ResNet_U-Net网络模型的计算机自动分割视网膜动静脉算法。首先,使用多光谱视网膜成像系统采集视网膜图像并制作数据集,其包含206张548 nm光谱的视网膜图像及其像素级标记的动静脉图像;然后,优化ResNet_U-Net网络模型中的多尺度特征提取模块和损失函数模块,并增加了通道注意力机制和后期处理方法来提高动静脉自动分类的准确性;最后,在数据集中随机抽取165张作为训练集,41张作为测试集进行了测试。实验表明,本研究所建立的深度学习模型能自动准确分割视网膜图像中的动静脉,准确率可达98.50%。In order to reduce the labor-and time-intensive burden of manually marking arteries and veins,an algorithm for automatic segmentation of retinal arteries and veins based on the ResNet_U-Net network model is proposed in this research.First,retinal images were acquired using a multispectral retinal imaging system and a dataset was made.The dataset contained 206 retinal fundus images at 548 nm wavelength and their pixel-level labels.Then,the multi-scale feature extraction module and loss function module in the ResNet_U-Net network model were optimized.And a channel attention mechanism and post-processing methods were added to improve the accuracy of automatic classification of arteries and veins.Finally,165 images were randomly selected from the dataset as the training set,and 41 images were tested as the test set.Experiments show that the deep learning model established in this study can automatically and accurately segment the arteries and veins in retinal images,with an accuracy rate of 98.50%.

关 键 词：深度学习 动静脉分割 注意力机制 多光谱成像系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TH7[自动化与计算机技术—计算机科学与技术]